Binary classification - false positive and false negative

Question

I have a doubt in the terminology adopted on below learning page for false negative and false positive -

https://learn.microsoft.com/en-us/training/modules/fundamentals-machine-learning/5-binary-classification

How is predicted_value=0 and actual_value=1 classified as false positive? This should be false negative. Same in case of what is specified as false negative on the learn web-page.

False positive is when model predicts presence of something, but in actual its not i.e. predicted_value=1 and actual_value=0.

I believe these values if toggled (i.e. if not represented correctly) will cause incorrect calculations for Recall, Precision and all other derived values from them.

Appreciate a feedback, in case I am missing something here.

Answer 1

You're correct. FN and FP have been transposed. We'll update the unit ASAP

Answer 2

An update has been pushed and should be live within a few days - thanks for calling this out.